Aakash Desai, Assistant Professor at UAB O’Neal Comprehensive Cancer Center, shared on LinkedIn:
“Artificial intelligence (AI) is making strides in mammography screening!
The PRAIM study, conducted across 12 sites in Germany, evaluated AI-supported double reading vs. standard double reading for breast cancer detection in 463,094 women.
Results:
– Higher cancer detection: AI group detected 6.7 cancers per 1,000 women—17.6% higher than the control group.
– Noninferior recall rate: AI group achieved a recall rate of 37.4/1,000 compared to 38.3/1,000 in the control group.
– Improved precision: Positive predictive value (PPV) of recalls was 17.9% with AI vs. 14.9% without AI.
What does this mean? AI-supported screening enhances breast cancer detection rates without increasing recalls, demonstrating its potential to improve screening outcomes in real-world settings.”
Authors: Nora Eisemann et al.
